Applied AI/ML Lead

As Applied AI/ML Lead within Commercial & Investment Bank with the Healthcare Provider team, you will lead the design, development, and production deployment of AI/ML solutions focused on image classification, text categorization, and data extraction from scanned TIF documents.

You will architect and implement computer vision pipelines leveraging CRNN architectures for document type identification, page-level categorization, and visual feature extraction.

Job responsibilities


* Lead the design, development, and production deployment of AI/ML solutions focused on image classification, text categorization, and data extraction from scanned TIF documents and evaluate and explore additional models and architectures to continuously improve classification accuracy, extraction quality, and processing efficiency.


* Drive the development and fine-tuning of models for document understanding, text categorization, named entity recognition, and semantic understanding and combine visual layout information, textual content, and spatial relationships to extract structured data from complex scanned documents, while enabling automated categorization and metadata tagging of OCR-extracted text.


* Lead the integration and optimization of OCR technology and generative AI capabilities into the document processing pipeline, ensuring high-accuracy text extraction from scanned TIF images across diverse document types, layouts, fonts, and quality levels.

Leverage Amazon Bedrock to explore foundation model capabilities for intelligent document understanding, classification, document summarization, and augmenting traditional extraction pipelines.


* Architect and implement scalable ML training and inference pipelines using AWS SageMaker, managing model training, hyperparameter tuning, distributed training for large vision models, and real-time/batch inference endpoint deployment.

Collaborate with software engineering teams to integrate trained models into Java/Python-based microservices deployed on AWS EKS, ensuring low-latency, high-throughput inference for production document processing workloads.


* Establish robust MLOps practices and annotation workflows, including model versioning, automated retraining triggers, A/B testing of model variants, drift detection on document distributions, and comprehensive performance monitoring dashboards and design and manage labeling strategies for training data, ensuring high-quality ground truth datasets for image classification, text categorization, and document extraction tasks.


* Build and manage a team of ML engineers and applied scientists, fostering a culture of experimentation, rapid prototyping, and rigorous evaluation of model performance against business KPIs.

Required qualifications, capabilities, and skills


* Bachelor's degree or MS or PhD in quantitative discipline, e.g.

Computer Science, Mathematics, Operations Research, Data Science.


* 7+ years of experience in applied ML/AI rol...